Applied statistics and its development by Sergey Aivazian

I had met Sergey Aivazian (hereafter referred to as SA), aka Sergey Artemievich (by the name and patronym, commonly used in Russian) in Moscow, circa 1975–1976. At the time, SA was already a well-known statistician, doctor of sciences, professor, as well as the head of the laboratory of probability and statistics in the Central Economics-Mathematical Institute (CEMI) of the Academy of Sciences. Meanwhile, as a physicist by education I could not find a job – instead, I started to work in economics as a mathematician/programmer who was supposed to develop some miracle statistical models to push and elevate the soviet economy. The first book I had read on the applied statistics was SA’s “Statistical investigation on dependencies” and I loved it very much. Later I had read SA with co-authors’ books on “Classification of multidimensional observations”, “Typology of consumption”, three volumes of “Applied Statistics”, attended the seminars on “Multidimensional statistical analysis and probability modeling of real processes” which SA held in CEMI for many years. With all this new knowledge, I began to employ various statistical modeling and predictions for practical aims.

In that “stagnation” epoch of 1970–1990, the soviet economics had been weakly functioning under such unclear slogans as “Economics should be economical”, but there was a belief that using mathematical and statistical modeling, computers and automatic systems of management could significantly improve the country’s economic growth. However, there were strict ideological regulations and suspicion that “western” influences are disastrous for soviet life. For instance, the econometrics models were still defined as a “bourgeois” science serving the capitalists in their suppression of working class (a description of that time is given in my book Boulevard of Moebius. Lulu, 2019). The linear programming and interpretation of its dual solution as “shadow prices” was considered as an anti-Marxist view, and it was forbidden for its founder Leonid Kantorovich to attend the reception of his Nobel prize in 1975. I remember the 100-th anniversary meeting in honor of the Central Statistical Bureau journal “Statistical Bulletin” (“Vestnik Statistiki” in Russian) where the guards of soviet economics criticized the methods of multivariate statistics as “formalism”. For instance, “Hilbert space” or even calculus was named harmful for national welfare – and SA had to defend those “modern” techniques as necessary and useful tools of statistical estimations.

In the 80s, when I had been preparing my Ph.D. thesis on mathematical methods at the Department of Economics, Moscow University (MGU), SA had become my adviser, after my first adviser Prof. Aron Yakovlevich Boyarski passed away. As the deputy director at CEMI and professor in MGU and other universities as well, SA was an extremely occupied and busy as a scientist and science coordinator. None the less, he was a really generous person I could always approach to discuss multivariate techniques, methods of index analysis, and whatever else I needed to clarify. Sometimes, I substituted for SA in lecturing for graduate students in MGU, and I was amazed how high his standards for teaching statistics were.

SA was always a key organizer in numerous meetings and conferences on applied statistics and probability, economics and econometrics, quality control in production, algorithms and software development, which were regularly held in various places of the USSR. For example, SA arranged them in Moscow and in Perm (the capital, and the Ural mountain, Russia), Tsakhkadzor near Erevan (Armenia), and Tartu near Tallinn (Estonia). I attended many conferences during the span of fifteen years, and presented multiple works there. SA had also suggested to publish my papers at various professional journals, such as “Industrial Laboratory” (“Zavodskaya Laboratoria”), “Economics and Mathematical Methods” (“Economica and Matematicheskie Metody”), and in the series of books of the selected works presented at the seminar in CEMI.

Many years later, in 2008, I served as the program chair for the marketing research section in the Joint Statistical Meeting (JSM) of the American Statistical Association. I asked SA to be an invited speaker in Denver, CO, at JSM attended by more than six thousand statisticians. Initially he had agreed and sent to me his resume in English, needed for presenting to JSM committee. However, he was so busy with his numerous works and obligations, that he changed his mind and could not spend time on such a vacation in the USA. However, I kept that his Curriculum Vita written by SA himself – and this document (slightly cleaned from some redundant detail) is given in the Appendix. It presents a valued list of the works and achievements of a prominent statistician of our time, a useful document for the history of sciences. Hard working, powerful, and energetic – Sergey Artemievich Aivazian always came across as an amazing person for anyone who ever met him.